Some years ago, in the genomic era, it was thought that all information of the cell was coded in the DNA molecule, meaning the code written with its four bases A,T C and G. But it has been shown that there is an extra layer of information which is not coded in the DNA, which consist of post-translational modifications that in part occur on histones, the proteins over which the DNA molecule is wrapped around to. These modifications, also known as the epigenetic code, regulate the accessibility of the DNA for the transcriptional machinery, they are heritable and also one of the major mechanisms that lead to the development of diseases.
Among these modifications, histone acetylation is a hallmark of chromatin that is accessible to the transcription machinery, meaning it is a “go transcription” sign. Acetylation on histones is controlled by ‘writer’ and ‘eraser’ enzymes: histone acetyltransferases (HATs) that produce or write the acetylation marks and histone deacetylases (HDACs) that erases them. But this ‘acetylation’ code has to be interpreted and this is done by effector ‘reader’ modules, named bromodomains. These are approximately 110 amino acid modules found in several chromatin-associated proteins and transcriptional regulators, that upon recognizing acetylation marks, translate the information within in the context of chromatin reorganization and transcriptional control.
Bromodomain-containing proteins (BCPs) are trouble when deregulated and their bromodomains appear to play important roles to disease mechanisms. This association between BCPs and disease has made scientists, from both the academia and the pharma world, rush to their benches in the last six years towards a new goal: the development of bromodomain inhibitors for drug discovery purposes and of course to understand the roles of these BCPs in a biological context. So far ten compounds blocking protein-protein interactions of a the BET subfamily of bromodomains have entered clinical trials. Nevertheless, the development of inhibitors as chemical probes of individual BCPs within the same family has remained a major challenge, due to lack of single target selectivity.
THE RISE AND SHINE OF BROMODOMAINS
In the 1990s, the first bromodomains inhibitors were discovered by scientists of the now Mitsubishi Tanabe Pharma. In their patent obtained in 1998, they described them as potential therapeutics for the treatment of inflammatory intestinal diseases. Later, in 2006, in a new patent obtained by the same company, these inhibitors were shown to inhibit CD28 co-stimulatory signals in T cells, making them potentially useful for suppression of rejection response in transplantation and treatment of autoimmune and allergic diseases. However, these molecules remained relatively unknown…. until 2010, when two independent groups reported small-molecule potent inhibitors of the BET bromodomains subfamily, called JQ1 and I-BET. That was when it all suddenly changed and the scientific world started to put their eyes and efforts in targeting protein interactions mediated by epigenetic readers of acetylation marks. But what made them so interesting and popular, is that they showed in vivo on-target activity in models of NUT midline carcinoma (a very aggressive cancer that has a mean overall survival of 6-9 months) and inflammation (conferring protection against lipopolysaccharide-induced endotoxic shock and bacteria-induced sepsis) respectively. It was these two groundbreaking discoveries that demonstrated the high ‘druggability’ of the bromodomain-acetylation interaction and that motivated further drug development efforts and the development of molecular probes to identify the function of bromodomains.
TARGETING THE BROMODOMAIN OF BRD9
In order to have different gene expression programs from a single genome, chromatin structure must be precisely regulated. Want to transcribe?, then loosen chromatin; want to avoid transcription? Then condense chromatin. This structural regulation is possible thanks to several chromatin-remodeling complexes, some of which uses the energy derived from ATP hydrolysis to disrupt histone-DNA contacts to control access of transcription factors to DNA. One of them is the 12-subunit complex SWI/SNF, which plays a crucial role in tumor suppression. In fact, nearly 20% of human cancers harbor mutations in one or more of the genes encoding SWI/SNF. Despite this role, the mammalian SWI/SNF complex has only recently received attention as a possible target for therapeutic inhibition, because only since 2013 it has been shown to be critical for the growth of genetically defined cancers.
The bromodomain–containing protein BRD9, one of SWI/SNF subunits has recently been shown to be required for the proliferation of acute myeloid leukemia cells, although its function within the SWI/SNF complex is not known. This elusive biological function has motivated the development of inhibitors against the bromodomains of this protein to identify the activities that drive disease pathology. So far only five BDR9 inhibitors have been published: LP-99, “compound 28”, I-BDR-9and the most recent ones BI-7273 and BI-9564. The main problem has been the high similarity between BDR9 and BDR7 bromodomains (~80% sequence homology), making the development of a selective ligand for just one of these bromodomains a very difficult task. While this could not be achieved by LP-99 nor “compound 28”, it was achieved by GlaxoSmithKline’s I-BDR9 with a 200-fold selectivity over BDR7 and greater than 700-fold selectivity over the BET subfamily members.
The newest inhibitors were described in a study published this year, where scientists at Boehringer Ingelheim in collaboration with SGC-Oxford and Cold Spring Harbor Laboratory, developed the inhibitors BI-7273 and BI-9654 using two parallel screening approaches: fragment-based screening and extensive structure-guided virtual screening. The compounds are potent, highly selective (>1000-fold) against BET subfamily and a panel of kinase and GPCR targets, block AML cancer cell and display efficacious antitumor activity in a xenograft model of human AML, providing the most high-quality chemical probes for BDR9 and BDR7 so far.
For this study three parallel biophysical assays were used to screen around 1700 compounds against BDR9, a differential scanning fluorimetry (DSF) assay, a surface plasmon resonance (SPR) assay, and a MicroScale Thermophoresis (MST) assay. The primary screening hits identified by MST was significantly higher than that with the other two techniques (124 for MST vs 36 for DSF and 45 for SPR), with only 10% overlap among the primary hits on the three technologies. 38 of the MST hits were validated by NMR, from which 29 were identified solely by this technology, and from which 14 co-crystal structures could be obtained. Thus this study emphasizes the significant value of persuing single-technology positives.
The progress achieved so far shows that it is possible to obtain inhibitors of bromodomains, particularly towards non-BET family members, such as BRD9. In this process, controlling selectivity against other families of bromodomains is a must, but also to achieve intra-family single target selectivity, which remains a challenge. This selectivity is very important for chemical probes, because they have to fulfill requirements that drugs do not necessarily do; they must have a defined mechanism of action and should be as selective as possible, ideally for a single target, while drugs must be safe and effective at treating disease.